THE EARTHQUAKE ADVANCE THEORY 195 higher beaches by McConnell in 1904, however, with mature forest hundreds of years old upon them, showing earlier lakes 300 feet deep, suggests that the glacier-dam explanation is correct and that there had also been still earlier advance and damming, perhaps contemporaneous with the earlier and greatest advance of the glaciers in Yakutat Bay, and elsewhere on the Pacific coast. Still earlier, but perhaps as a part of the same great advance, the glaciers had been so extended that the sites of these lakes were completely ice-covered. The north side of the St. Elias Range, therefore, serves to show the same glacial history as the south side:—(1) former great expansion of glaciers (perhaps with the 800 foot, forest-covered beaches associated with stages of ice-barrier lakes during retreat); (2) retreat of glaciers (since we cannot assume that the lakes stood at the 150 foot stages long enough for mature forest growth on the beaches above); (3) readvance a century or thereabouts ago with the formation of the 150 foot and lower beaches, since which there has been continuous retreat. As shown by Wright1 and by Reid,2 Glacier Bay also furnishes evidence of a similar succession of recent events to that in Yakutat Bay and the north side of the St. Elias Range, and essentially contemporaneously with it. We may therefore assume that the last great expansion of the glaciers of the east end of the St. Elias Range was due to a cause which affected a mountain area with a northwest-southeast axis of at least 150 miles and a width of over 50 miles. But in Prince William Sound we find no evidence of recent glacier expansion. There was a period of great early expansion corresponding to that during which the Yakutat Bay glaciers reached their greatest extent, completely filling the inlet; but this period was centuries ago. Mature forest now extends up to the fronts of some of the glaciers of Prince William Sound and nearly up to the fronts of others. They are, therefore, now almost as greatly expanded as at any time within a century or two; and Columbia Glacier is greater than it has been for a long time, for both at its sides and front it is encroaching on the mature forest. It is possible that these glaciers are now in a stage of advance comparable to that of the Yakutat Bay glaciers a century or more ago. Three possible explanations suggest themselves to account for this recent pronounced, but brief, advance of the Yakutat Bay glaciers and those in the two neighboring regions cited. The first of these is the diastrophic hypothesis, that uplift took place and the mountains were raised so much higher that the glaciers were thereby given greater snow supply. This hypothesis fails in one vital respect; by it we cannot account for the brevity of the advance and for the subsequent recession. It may, therefore, be dismissed as inadequate to account for the phenomena, as is the complementary hypothesis of submergence proposed by Reid to account for the retreat of the Glacier Bay ice tongues. The second hypothesis is the vague one of climatic change, making no attempt to explain why it came, how it operated, or what its exact nature was, but merely for some unknown reason an increase in snowfall, or change in temperature, came to these mountains and that ultimately the glaciers flowing from them responded by advancing; then came diminution in snowfall, or increase in temperature, and accompanying recession. Lacking knowledge of the amount and nature of the snowfall and the temperature of the critical parts of the Alaskan coast, as we do, it is impossible to apply to this theory the i Wright, G. F., Ice Age in North America, New York, 1891, pp. 51^67. > Reid, H. R, Studies of Muir Glacier, Alaska, Nat. Geog. Mag., Vol. IV, 1892, pp. 84-40; Glacier Bay and Its Glaciers, 16th Ann. Rept., U. S. Geol. Survey, Fart I, 1896, pp. 438-440.